Armin Waibel wrote:
thank you very much for your fast response :)
for your solution (e.g. -Xms128m -Xmx256m) I already used in my jvm
be have you more information for performance OJB + and very great Data base
or have you information for problem use java JVM whith OS (debian linux)
Thanks ...
Sorry no for both.
But I think there shouldn't be a problem in general. But if you perform a query with a result of millions of objects you will get problems with all O/R mapper ;-)
OJB provide various methods to avoid materialisation of large result sets:
- lazy loading for references
- customizing collection queries
http://db.apache.org/ojb/docu/guides/advanced-technique.html#Customizing+collection+queries
- report queries
http://db.apache.org/ojb/docu/guides/query.html#Report+Queries
- set start/end index on Query (#setStartAtIndex(...))
- use a cache that allow gc of cached objects (use SoftReference) when memory runs low (default ObjecrCache do so)
and a Profiler will help you determine where the problem lies (e.g. of what class you get millions of objects or similar). Since you're under Linux you might for instance want to try JMP which is free:
http://www.khelekore.org/jmp/
Tom
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
